Todd Lipcon has posted comments on this change. Change subject: KUDU-1856: always truncate containers when they get full ......................................................................
Patch Set 1: (3 comments) http://gerrit.cloudera.org:8080/#/c/5852/1//COMMIT_MSG Commit Message: PS1, Line 18: truncate. Instead, all full containers are truncated unconditionally. The : assumption here is that a no-op truncate should be free. I think it's worth validating this assumption (eg by timing startup on one of these hosts that has 10k+ full containers, after a drop_cache=3). It would suck to find out only upon release that this is causing a journal fsync per truncate call or something. http://gerrit.cloudera.org:8080/#/c/5852/1/src/kudu/fs/block_manager-test.cc File src/kudu/fs/block_manager-test.cc: Line 1080: ASSERT_OK(env_->GetFileSizeOnDisk(fname, &size_after_preallocate)); have you tried this new test on ext4 on el6? I'm not sure whether fallocate KEEP_SIZE works on old ext4, would be good to make sure we don't have a surprise test failure in that env http://gerrit.cloudera.org:8080/#/c/5852/1/src/kudu/util/env_posix.cc File src/kudu/util/env_posix.cc: Line 137: CHECK_EQ(0, mode); I thought it was mode, 0 -- To view, visit http://gerrit.cloudera.org:8080/5852 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-MessageType: comment Gerrit-Change-Id: Ic959c59489a08f92efa2df5c85b22e56740f1346 Gerrit-PatchSet: 1 Gerrit-Project: kudu Gerrit-Branch: master Gerrit-Owner: Adar Dembo <a...@cloudera.com> Gerrit-Reviewer: Kudu Jenkins Gerrit-Reviewer: Todd Lipcon <t...@apache.org> Gerrit-HasComments: Yes